The long-term objective of the proposed research is to study the immunologic relationships among various moust mammary tumor viruses by cross protection tests in mice and to determine, through similar tests, possible relationships between mouse and human mammary tumors. We will test the vaccines prepared from mouse mammary tumors, purified MMTV's, and purified subviral components, gp55 (or gp50) and p28, in a relatively rapid screening system, using mice which ordinarily do not express high incidence of spontaneous disease but which are exquisitely sensitive to challenge by infectious MMTV's of the S ad P types. These studies will serve as a guide for vaccination in mice which express spontaneous disease at high incidence and which require longer holding periods for analysis. Using the same strategy, we will study the relationship between the diseases in mice and mammary carcinoma in humans. Human tissues will be obtained from pre- and post menopausal women demonstrating either infiltrating ductal carcinoma or lobular carcinoma in situ, as primary tissue or cell culture. In addition extracts will be prepared from primary human tumors. In all studies, appropriate controls will be used as immunogens including non-mammary tumors, benign lesions, non-B type viruses and subviral components of the latter. Regardless of vaccines used we will ask whether there is a reduction of or a delay in the appearance of: MMTV antigens which appear in milk of lactating mice prior to tumor formation; early or late mammary tumors which appear as a result of horizontally transmitted milk virus (early), or vertically transmitted endogenous virus (late); nodules (and tumors) in those strains of mice wherein "L type disease" is active. We will ask whether the inhibition or delay which might be detected is virus "subgroup" specific (e.g. S, L, P type) or virus and mouse strain specific (e.g., RIII, GR, C3H, A).